Fundamentals of GIS

OVERVIEW OF GISG I S Geographic Information Systems An INTEGRATED SYSTEM of COMPUTER HARDWARE and SOFTWARE coupled with PROCEDURES and a HUMAN ANALYST which together support the CAPTURE,

EMPHASIZING GIS USE FOR NATURAL RESOURCE MANAGEMENT

David P Lusch, Ph.D

Senior Research Specialist

Center For Remote Sensing and Geographic Information Science

Michigan State University

November, 1999

OVERVIEW OF GIS

G I S

Geographic Information Systems

An INTEGRATED SYSTEM of COMPUTER HARDWARE and SOFTWARE

coupled with PROCEDURES and a HUMAN ANALYST which together

support the CAPTURE, MANAGEMENT, MANIPULATION, ANALYSIS,

MODELLING, and DISPLAY of SPATIALLY REFERENCED DATA

G I S Capabilities:

q QUERY FOR LOCATION

"Show me all the countries of South America that have a population greater than 20,000,000 "

q QUERY FOR CONDITION

"Display the population of each country I point to on the map."

q TREND ANALYSIS

"Show me where the census blocks are that have experienced more than a 50% population change between 1980 and 1990."

DATA TYPES

Point = Position, no area

RASTER VECTOR

Point = 1 cell

Polygon = Area and perimeter Polygon = Group of contiguous

cells joined at edges or corners

Line = Length, no width Line = Multiple cells joined at

edges or corners, usually with only 1 or 2 neighbors Geometric Classes of Data

Raster

VECTOR STRUCTURE

Advantages

q Good representation of the landscape being mapped

q Topology can be completely described, including network

linkages

q Great looking graphics ("Looks like a map is supposed to")

q Generalization of the graphics is possible while still maintaining

the great look ("What the map reader doesn't know won't hurt

them")

RASTER STRUCTURE

Advantages

q Overlaying maps is easy and "perfect" (i.e no possibility of

sliver polygons developing since all raster cell borders are

coincident

q Integration of remotely sensed imagery (satellite images or

scanned airphotos) is straight-forward

q A huge variety of complex spatial analyses are supported

q Software is generally cheaper and easier to learn compared to

vector GISs

TOPOLOGY

Geometrical relationships between spatial objects (Points, Lines, and Areas), such as adjacency, that are not altered by distortion, as long as the

surface is not torn

Example of "Built" Topology (from Arc/Info)

Arc Left Right From To

ID Poly Poly Node Node

A Classification of GIS Functions

q Analysis of Spatial and Attribute Data

- Non-spatial analyses

Attribute query and display Map retrieval and display Attribute classification Map measurements (distance, direction, area, etc.)

- Spatial analyses

Overlay operations Neighborhood functions

Distance and Connectivity functions

Contiguity measures Proximity analysis Network analysis Spread functions Seek operands Intervisibility analysis Solar illumination calculation Perspective view

Search operations

Line-in-polygon; Point-in-polygon Topographic functions

Thiessen polygons Interpolation

Contour generation

This classification has been adapted from:

Aronoff, Stan 1989 Geographic Information Systems: A Management

Approach Ottawa, Ontario, Canada: WDL Publications 294p.

A Classification of GIS Functions

q Maintenance of Attribute Data

Format conversions Database error checking Database editing

q Maintenance of Spatial Data

Format conversions Geometric transformations Projection conversions Conflation

Edge matching Editing of graphic elements Line coordinate thinning

q Output functions

Map annotation Text labels Texture patterns and line styles Graphic symbols

Plotting Printing (laser printers, color inkjet printers, etc.)

This classification has been adapted from:

Aronoff, Stan 1989 Geographic Information Systems: A Management

Approach Ottawa, Ontario, Canada: WDL Publications 294p.

Vector and Raster

Vector and Raster

Vector and Raster

Spatial Analyses

Vector and Raster

Raster Only

Vector and Raster

Vector and Raster

Vector and Raster

All locations in the coverage or grid

are evaluated The results extend

to the spatial limits of the input maps.

Vector and Raster

All locations in the

coverage or grid

are not necessarily

evaluated The results

are constrained to the

spatial nature of the

reference map.

Vector Only

Note: Most relational database management systems support a "SWAP"

function which selects the currently unselected items in the database In

the example above, the "SWAP" function would return the white areas after the hatched areas had initially been selected.

Vector and Raster

Simple (i.e unweighted)PROXIMITY SURFACE

Raster Only

Raster Only

Least-Cost Pathway Analyses

q Create a FRICTION map from one or more

existingcoverages

q Create a COST SURFACE map by executing

the PROXIMITY analysis WEIGHTED BY the

FRICTION map

from one or more starter entities (points,

lines or areas) to a destination It will FIND

the one (or more) routes that ACCUMULATE

the least cost.

Base cost x 4

Base cost x 1000 Base cost x 5 Base cost

Cost Surface (Proximity x Friction)

Least-Cost

Pathway

Vector and Raster

SPATIAL ANALYSES

Basic Functional Classes

q Reclassifying Maps

q Overlaying Maps

q Measuring Distance and Connectivity

q Characterizing Neighborhoods Raster Only

WINDOW OPERATIONS

Slope Slope Aspect Maximum, Minimum Mean, Median, Mode Standard Deviation Majority, Minority Total Count Diversity

Large number of Spatial Pattern

or Texture Indexes

(e.g Dominance Index, Relative Richness, etc.)

Raster Only

Each of the following measures are calculated within a 3 x 3 pixel window which systematically roves

throught the data set The outcome calculation is assigned to the center cell in the window (in the output file), then the window moves over one pixel along a row and recalculates a new value for that center- pixel location, etc.

Relative Richness = n / (nmax) x 100

where n = number of different classes present

from: Turner, M.G 1989 Landscape Ecology: The Effect of Pattern on Process, Annu Rev Ecol Syst., 20, 171-197.

Examples of Neighborhood Operations to Determine

Spatial Pattern

Relative RichnessDominance Index